Background:
Order sets in electronic medical records (EMR) have shown to reduce variation in clinical practice, reduce error, and increase the quality of care. It is important to regularly evaluate appropriateness of existing order sets after implementation to optimize care and patient safety. Fecal occult blood test (FOBT) has been traditionally used for colon cancer screening. At our institution, FOBT was included in the IV heparin order; the reason for this has been difficult to discover. Our study evaluates the frequencies of FOBT ordered as part of the order set, the results and their documentation. We also explored the effect of the positive results in the clinical management along with rates of gastrointestinal bleeding.
Methods:
A retrospective study was conducted among 898 adult hospitalized patients initiated on the IV heparin order set our academic center during a six month period. Demographics, hospital admission information, diagnosis, discharge disposition, laboratory data, and indication for the heparin drip were collected.
Results:
Among 898 patients whom heparin drip order set was initiated, only 14.3% (128) patients had FOBT sent. Of these 128 patients, 33 (26%) were found to be positive. Only 36% positive results were documented by any provider during the hospitalization. An occult blood test was sent on 7 of these patients (21%) even though they had obvious signs of bleeding. In two patients (6%), the heparin drip stopped as a direct result of the positive hemoccult test. Gastroenterology was consulted on 10 patients (30%) who were OB+ and 7 patients underwent endoscopy. Among the 33 patients with positive FOBT, 13 were discharged home; almost half of them, (n=5) were readmitted to the hospital within 90 days with a diagnosis of GI bleed.
Conclusions:
Regular review of order sets to evaluate the utility and effectiveness of these sets in the clinical management is necessary. Careful considerations are needed to update these order set to improve the quality of care and to reduce the adverse consequences. In this review of practice, we saw that the test in question was not secured most of the time, and the results rarely changed management.